Furilabs Linux Phone
furilabs.com
furilabs.com
I am definitely interested, especially if the team working on it is as active as it seems they are. Definitely one of the biggest bummers with PinePhone and PinePhone Pro was realizing that the community was largely on its own; that could work well, but in practice progress has been slow and painful and I stopped paying attention.
I would really like to see more projects doing cellular Linux devices in general. It doesn't need to be a candybar smartphone, if someone can jam a cellular modem into a MicroPC I'm sold at any price I can afford.
I am obviously a little biased, but this is probably the first Linux phone that people can actually use. Someone in our community switched from iPhone to this without much of an issue.
0: https://wiki.postmarketos.org/wiki/OnePlus_6_(oneplus-enchil...
Since these days something like >90% of people is on WhatsApp, it’s beyond quintessential to have a good WhatsApp client.
There is now a law from the EU, I think it is the DMA, which could change things for people living in the EU regarding interoperability. I haven't read anything about practical follow-ups yet.
By the way, only releasing source code as a blueprint might be fine, that happened before. I might remember it wrong, but it might be lame, the mp3 encoder.
Probably not very detectable for Meta since it uses a wrapped webview, but I've been using ZapZap on Linux for a good while now and I haven't seen a ban yet.
Interesting, my naive understanding is that push notifications won't work and battery life should be greatly affected, how do things work for you?
[2]: In 2020, Facebook shutted down the NYU Ad Observatory by shutting down the Facebook accounts of Cybersecurity for Democracy team members Damon McCoy and Laura Edelson.
[1]: https://12challenges.substack.com/p/how-to-deal-with-receivi... [2]: https://cyber.nyu.edu/2021/08/21/facebook-disables-ad-observ...
In the end it also boils down to what devices the BankID app providers are willing to support, I have a hard time seeing anything but iOS or Android devices being supported in the near future, Esp as Swedish BankID's now also requires NFC support to read the local police issued ID cards (had to get a new testing-device just due to this requirement).
Note: BankID is the name of personal identity apps that support authentication and signatures in Sweden, Norway and Finland, the authentication is used to access a myriad of both public and private sites like tax office, unemployment, healthcare and gyms. The signatures done via the apps are generally accepted to be of as good legal standing as a signed paper.
On the positive side in Finland you can use SIM Toolkit for legally sanctioned 2FA (mobiilivarmenne). That should be much easier to implement without having Google involved.
I guess it's the opposite to what happened in Sweden where Telia was the only mobile provider that had an identity solution (that had any uptake), but afaik gave up due to the omnipresence of BankID that is a streamlined system supported by all the major banks.
As a private sector application developer, would one integrate with MV, the suomi.fi portal or is it a fragmented system?
Whether the banks have any common backend/provider or you need to deal with all of them sperately I don't know.
Also typing this on the ultra-cheap hardware made for the Indian market nearly a decade ago. Well, admittedly they have already announced to discontinue support and I never dared to carry it as my primary phone.
Now the last weeks I was forced to use Android as my primary phone. Really a suprise what crappy piecemeal Android is compared to SailfishOS made by a tiny company struggling economically forever.
I'm actually using GNU/Linux phones since 2008. Never relied on Halium.
Still, it's something!
Shift6mq is supposedly close, with OnePlus6(T) its more difficult due to I think the way the sensor is connected.
Currently, the Pixel 3a is the most promising device in terms of Camera Support on a Qualcomm device with mainline kernel from what I can gather.
While Nokia could maintain their own Linux kernel that's impossible for a small vendor/team. There is no realistic alternative to taking the Android kernel provided by the SoC vendor.
I'm sick of big companies who removed the word Consent from their dictionaries. I want something I can own and trust again. Being able to run android apps would be great, but given how closed source Android has become, I doubt that apps run on an open system would be reliable.
If second market also counts then there are a few more options available - mostly former Android phones where hardware support is spotty though, so do your research first. That said, everyone's needs are different and what's there can already fulfill the needs of plenty of people, so it's a matter of figuring out whether you in particular are in this group yet or not.
It's a little weird that there are very few details on this thing. Notably, one of my #1 concerns when looking for a new phone today is a good camera. And all I can find on the camera is "it's 50MP!!! " But how do actual, non-photoshopped pictures look? Does it do okay in low-light? Does it support any sort of optical zoom?
Their website is currently as slow as a frozen snail (which, if they can't run a basic web server, leads to a lot of doubt they can maintain a long-term Linux distro). So I couldn't click around much, but I didn't find anything really informative.
I tried looking at YouTube videos and only found some weird official videos of a person's hand putting the phone in a sink. Like... Yes, it has been expected for the last 8ish years that my phone will be waterproof, that's not an amazing achievement.
It also seems to be "weird" Linux, like Maemo was for the N900. I don't want weird, custom-built stuff that will be forgotten about in 12 months. Just give me a standard Debain/Fedora/Whatever and an unlocked bootloader to reinstall with what I want.
IMHO they should really look at what Valve did with the steamdeck and SteamOS, It's a custom Linux that works well, but, it's also standard hardware with an open bootloader. There are dozens of SteamOS alternatives.
I wouldn't say it's a "weird Linux as it's mostly Debian and it's fully rebuildable by the user from the open source scripts[1]. I would likewise love a phone that was just mainline Fedora, but given many of the current challenges around ARM devices with linux, I fully understand why a "custom-built" OS is needed. As long as it's as light-weight as possible and is open source/rebuildable by the user, I think it's a good compromise given the realities around the platform.
Maemo was "mostly Debian" as well, it was one of the big selling points. Except Maemo was quickly abandoned by Nokia and left unmaintained. To further complicate matters Maemo required arcane tools to build it and install it to a device. Custom community kernels quickly became standard for enabling basic functionality which Nokia should have provided, but, didn't. However, the custom kernels were also frequently left unpatched and abandoned because they were being built and maintained by random internet users who found other hobbies.
I think the idea is awesome and I am excited if not skeptical. I hope Furi manages to build a healthy community around this. And doesn't go the path of so many previous Linux phones.
I also completely agree that we should have a camera showcase. We've been talking about the idea of letting customers upload their shots to some sort of gallery so people can see how the real world looks through the camera's lens. What I can tell you though is that pictures look pretty darn good, and low light is great owing to the f/1.57 aperture. No optical zoom for this model unfortunately.
I'm not sure the website being slow speaks to our capability of maintaining a distro, especially when you consider these two things are done by different people. Sure, I'd love to work on the website and make it faster. But I'm busy working on the actual phone.
The video situation will improve as reviews come in. As for just giving you an unlocked bootloader and regular Linux... sure, I can do that. But you're not going to have audio, hardware acceleration, camera, etc. postmarketOS is a good example of this - despite years of tireless work, it's just really hard to get a regular distro working on a phone without leveraging Android drivers, at least for now.
Does this mean you are not working on mainlining this device? I'll take the slow and steady pmOS approach over sketchy Android shims. That's what ruined the Planet Computers Gemini PDA for me. A temporary solution is often surprisingly permanent.
It's probably all they know about the camera. Not even Apple make their own cameras. It's always someone else's (customized)product. If they're outsourcing hardware, they might not even know the customer code assigned to the contracted OEM.
> I don't want weird, custom-built stuff that will be forgotten about in 12 months
Then none of hardware and many of software features will work. There are no in-box standard implementations for e.g. phone call in GNU/Linux, so it'll have to be supplied in the form of a custom preconfigured distro with bunch of free preinstalled apps.
The fact that product necessitating free preinstalled apps to function leads to needs to supply OS as flavors of your own weird unsupported distro, like N900-specifix Maemo or Steam Deck OLED-specific SteamOS images.
You aren't going to like being instructed how to install and patch a SIP softphone just to make calls on North American carriers with vague hints for Canadian specifics. It's a phone, it'll be lame if it doesn't make calls out of the box. And there comes the 3 years old $DISTRO-$HWREV-$VERSION.img that instantly shows antenna bars right on the setup screen.
Actually I think it's unfair to call Maemo a weird unsupported thing. It didn't seem that far from bone stock desktop Linux.
SFOS is somewhat getting where Harmattan was before Nokia imploded. It's already better in some ways.
https://github.com/FuriLabs/issue-tracker/issues
I think it's fair to say that whoever buys this is in for a bumpy ride. I mean, that is to be expected, however, I think they should be more honest about the current state, like for instance PostmarketOS does at https://postmarketos.org/state/
If SMS doesn't work, I have serious concerns about everything else...
Whenever I've tried to find a touch-optimized Linux shell, I've come up short. As has been discussed on this forum in the past, most of the attempts are understaffed side projects that have thousands of manhours of work ahead of them.
I'd like to know what shell FuriOS is based on, and what they're doing to make it usable. Even a demo video would go a long way!
Maybe one day I'll take a stab at fixing their package.
As someone unfamiliar with Furilabs, it wouldn't occur to me to go looking for a dev blog to see if you've written anything about the shell. I feel like that's information you'd want to link from the marketing page (even if the details are on your blog).
Fair enough! Noted.
It's on HN here:
I believe the GPD Pocket is also x86_64.
Fast, Good, and Cheap. Performant is just a synonym for fast. Is it any good?
Also call me a miser, but $499 is not "cheap". I recently bought a new Nokia G42 with the same camera and storage for <€150
To get under €150, you need volume and bundling.
edit: looks like same HW as "Gigaset GX6" https://www.gigaset.com/hq_en/gigaset-gx6/
seems pretty favorable, keep in mind that it's already two years old. it's also a brick at 280g, almost double a pixel 5, otherwise pretty similar
What's the difference between fast and "performant"?
Although more likely just weird marketing.
That said, it is awesome they are making this phone, and even if you don't use it, more choices are better for everyone.
The Pine64 PinePhone is $150-$200, with similar hardware provenance, but various open source projects that might at least give better software longevity and software trustworthiness. https://pine64.com/product-category/pinephone/
An additional consideration: as much as we like and respect our global colleagues, unfortunately I could see both hardware brands getting banned in some locales and by some organizations, like some other telecom brands have been banned. A phone that can't be used as a phone has little value, no matter how open it is.
For somewhat different provenance, the Purism Librem 5 is coming in at $800+, which is too high a high price point for most open source enthusiast contributors. https://shop.puri.sm/shop/librem-5/
And Purism's made-in-USA version, the Liberty Phone, is at $2K+. I guess maybe government or enterprise sales? https://shop.puri.sm/shop/liberty-phone/
I'm thinking that better for the Western open source developers and non-wealthy enthusiasts would be something more palatable to Western governments, and usable as a daily driver, like Librem 5, but priced somewhere in (guessing) $200-$400.
no not at all Pine64 hardware is pretty crap (performance wise, very normal independent of the OS at that price point)
Like you are comparing things like 3GB LPDDR3 SDRAM (Pine64) to 6GB LPDDR4X which is world apart (not only does it move from a "very tight on memory" to "somewhat acceptable memory" (very big deal) it also is much faster (~2x).
Similar on the Pine64 you have 4 old little cores here you have 2 big + 6 little cores and all more modern (end even just the little cores are faster when comparing one core with another, ignoring that you have 2 more little cores and 2 big cores).
Storage moves from a quite tight 32 GiB (through I guess if you only do stuff like mail it might be enough) to a likely more then enough 128GiB (also faster).
Noticeable faster GPU too, I guess.
And way better camera, at least by by numbers.
So not comparable at all.
But what matters most is not the absolute differences but moving it from a "so tight on RAM,CPU,GPU,Storage that you run all times into problems" to a "enough for all common tasks". I.e. from you need micro optimized apps for it to work nice to yes it works if your app is imperfect territory. And that is a pretty big deal.
But also lets be realistic. Atm. _any_ Linux Phone is a pure enthusiast product balancing various problems (price, small production lines, only android diver problems, proprietary blob problems etc.). And many of the enthusiast earn enough money to afford a 500$ toy, through maybe just rarely and with thinking twice about it.
I meant provenance as in geopolitical. Later in the comment alludes to this more, and one way it can be a showstopper.
Sorry I was being vague, but it's awkward to mention, and unfortunate that there is ever any potential conflict.
I honestly don't know of an "easy" way to do that. If you have enough time on your hands you could decompile it and take a look. You could also try running it on a spare phone with a ROM that doesn't contain Google Play Services (or where it's been disabled). However, SafetyNet isn't the only line of defense. These apps often have their own root checks, which doesn't pose a problem for us because the Android image isn't rooted. (But it runs inside LXC, and you can attach a root shell into it.)
Another thing they don't like sometimes is the system build fingerprint not being recognized. We have an overlay(fs) system that allows users to customize everything inside the Android container without the hassle of rebuilding the system images or having to stay on the ball when we update things. It's not hard then to pick up the build fingerprint line from a "trusted" device and just drop it in build.prop.
All of this is a bit of a hassle, to put it mildly. But if we try to talk to any bank or contactless payment provider right now, they'll tell us to kick rocks. I guess it's a matter of time until we grow or a very smart person figures out a very clever hack. :)
It's much much slower (and ugly, which doesn't help)
What's the difference between fast and performant?
For example, a car can have 1,000hp, but utilizes a large turbo which takes a long time to spool up, leading to poor performance. It may also have poor handling characteristics.
On the other hand, let's use a car like a Mazda Miata, which only has 181hp. It is not fast, but it is an incredibly performant car, with exceptional handling and driving dynamics.
Speed and performance are not directly correlated, and are independent metrics. Performance is more of an indication of the overall dynamics of action, intention and results.
Don't parse "covfefe", discard it.
280 grams is a lot even for today's phones. 12 mm is half a inch. Very thick.
The other dimensions are 6.7 x 3.2 inches.
Ok, this might be a phone for people that want to prove a point but it's too cumbersome. How about starting with something reasonable and fit it with only what fits within those bounds? An example: 160 x 80 x 9 and 200 g. That would be a big phone but not bigger than many phones on sale.
There's an undercurrent of despair in your comment. Don't do it to yourself, you only get one brain. Help us available. DM me if you want.
Because in that universe your comment would be about how awful the specs were and how tiny the battery is. I'm reasonably confident that price-size-performance is a "pick at most 2" situation.
You want 1/3 of thickness and 53ml in volume and 80g in weight just removed because methinks that'd be much better??? A phone isn't a steak cut, you... can't do that. You can if you knew beforehand that there are suitable replacements for major components like displays and batteries, if you don't, you just take what hardware engineers give you. What are you even thinking.
LineageOS
CalyxOS
GrapheneOS
/e/OS
Here is my personal favorite Audio Player base device, best I found so far:
In Android, headphone remote works out of the box... that's why I would like to use one. There is the HiBy M300, which in my opinion would be worth having a custom rom :-) Indeed there seems to be efforts.
https://github.com/reindex-ot/hiby_digital_m300
https://www.head-fi.org/threads/hiby-digital-m300-android-di...
This one looks pretty good!
It even has a feature that should make it a great migration away from android:
From an idealist standpoint, it shouldn't be done that way by taking away rights from users, but trivializing it as how specifically Google wants you to do it isn't accurate either.
Edit: In a different comment one of the devs clarified that it does have microG, so quite a few apps will actually work.[1]
With currently available payment standards, it's like asking for a live credit card with someone else's private key and an open debug port, all at the same time. Things don't work that way.
If the King of England walked into the Bank of England and wrote a hand-written bond for GBP1k with a bic pen on back of copier paper, they'll take it. Electronic platforms has nothing to do with that. That's why it's stupid to keep asking if banking apps would work.
what has gone wrong in your brain that you think that?
I'm referring to the pugnacious tone, but the certainty in the absence of evidence is also a bit gauche.
So you know, chazza banks with coutts. I reckon he could draw much much more, over the phone or in person, and wouldn't be put on hold or asked security questions. If you also happen to be the head of state of several nations and have your face on the coin of the realm, and your experience is contrary, then please do rebut my assertion.
Because that's the only question that matters. I have just a few apps installed, some of them are easily replaced, like TOTP apps, others are impossible to replace and without them I might as well get a feature phone.
I'm down to one app that I really do need, MobilePay, but it's a payment app, heavily controlled by the banks. That is not going to be installable on CalyxOS or anything without either Apples AppStore or the Google Play Store.
Phones doesn't matter, they are mostly the same, the operating systems are interesting, but without a widely supported app store, it does not matter, the phone will be useless for the majority of people. More and more people are using their phone as their only computing device, they need to have access to banking apps, mobile payment solution and government application, none of those are not going to exist on some random Linux phone.
The answer to the latter is and will continue to be an instant negative for quite a while, and it's pointless question to make other than as a really indirect nudging for big bank(lol).
Can we get a Linux smartphone that works before worrying about nonessential use cases? One the powers-that-be will never support until step zero is shipped at volume.
I can't get their timeline page to load correctly: https://furilabs.com/timeline/
How's the performance compared to a Linux vm running on a pixel 9 with kvm?
Android 15 recently released, which brought support for graphical acceleration inside the vm, so that's one competitor for the Linux phone.
> ...then the next planned feature is wireless display support. At present, you can't dock the phone and use it as a PC with a keyboard and mouse, which is a shame. Apparently, the IO lines needed for a wired dock were used for the macro camera.
https://www.theregister.com/2024/06/26/furi_phone_flx1_debia...
If in any doubt about whether to buy this, look at Planet Computers offerings and the quality of their Linux build. They have been offering Mediatek-based Linux phones for several years. Spoiler: garbage with ancient proprietary kernel, thanks to Mediatek
Definitely cool though and worthy of some attention.
And yes I like many others here have been trying Linux phones for 10+ years, since N900.
Your phone just has to work, it can't crash because you get a call, or hide buttons because the UI is confused by orientation. IT MUST WORK. So that's why I wonder why people don't just use Android.
I think a much more sustainable approach is to focus on existing Linux distros and get them more usable on phones.
I don't know what you mean by "look and feel of early android" either, as AOSP based ROMs look just like Google Pixel in the settings, quick dropdowns and dialogue popups.
But sure, LOS proves it can be done.
To be clear I have nothing against lineage, and I'm a Graphene OS user myself, but these projects add a lot of value on top of raw AOSP.
It only benefits the open source market in the long run because a working device will garner wider adoption.
That's the only thing stopping this from being a full computer.
I have a One Plus 12, which has HDMI out, but it's like they didn't feel like finishing it. It's not really usable, it doesn't have any controls like Dex does.
Because there debian, android? What is the bottom of it?
Let's not forget to also get it all audited, too.
Couldn't find too many datasheets nor other documentation unfortunately, though it might because of the unusable piece of crap website, which made me rage-close it.
[0]: just noticed: inFURIating, as in Furi labs, haha
> Furi is pronounced “Fury”. FuriOS is pronounced “Furious”. We simply couldn’t afford the Fury part of the domain names.
$800 now:
https://shop.puri.sm/shop/librem-5/
But yes the "nice" one aka Liberty Phone is still $2000, unfortunately.
The starlite tablet from star labs is quite good. Wish all these tiny companies could work together somehow.
One single integrated experience. I think everyone wants this. We have the technology we have the demand, it's just companies who make the high quality phones want to completely lock down the market.
You would think the free market would produce a company that would tackle this issue but barrier to entry is waay to high so essentially we have small companies trying to get into this space but they are only creating sub-par products because they can't make something as good as an iphone.
It's basically a legal monopoly. Companies can't get into it because the technology and capital required are way to high and only super mega corps like samsung and apple can pull it off...
Closest thing I've seen to this product I'm looking for is the steam deck. I would buy a steam deck phone if it had the quality of a pixel/iphone device.
That's where I have a hard time. I would pay that kind of money, but I would need something well polished and fully capable of being a "daily driver." I think many people are in the same place I am, and thus we have a real chicken and egg problem here.
I mean the technology to do this is already here. If apple or samsung wanted to do this... they can.
Right now I use an iphone, but if samsung made a phone that felt like say linux, windows or macos when in laptop mode... I would switch off iphone in a heartbeat.
And i mean it has to feel like macOS. None of that bloat is acceptable. SteamOS actually pulls this off but in a gaming form factor.
I think it really depends on what precisely you mean here. I don't think it's currently possible to get the same performance of my M3 Max MacBook Pro in the form factor of a phone.
I suspect that every single thing you like about macOS is something I consider bloat (or would if that was the only related word I was allowed to use).
My first task on my recently acquired M3 MBP was to remove/hide/disable as much of macOS as I possibly could; only then can I use it as a productive development environment.
Continuum wasn't exactly full convergence, but it was kind of close. As I recall, they had a desktop dock and a laptop like dock, that you could pair your phone to (wireless or wired).
Microsoft was big on their 'universal windows app' concept at the time, where universal meant actually only ran on a small fraction of windows devices. sigh
This is insane to me. Like not just no. Hell no. I want my phone to be a gadget, like a watch, not a whole ass computer. I want my tablet to be its own thing from my laptop, which is different from my PC(s). I absolutely, positively do NOT want general purpose computing to be the typical paradigm. I don’t want, need, or even like the idea of my phone, that I typically use like a wallet, to be a general purpose computation tool. Lock that shit down, please!
But what's the point in owning three devices when you can own one doing everything? How about in addition an mp3 player, a camera and a flashlight, too?
I mean the top post is a ~200USD “smart” pomodoro timer. I’m not even close to that.
I use my phone for communication, quick photos/ notes and sound. It carries my cards. I want it to be completely locked down because it gets used in public and could be stolen.
I use my laptop for personal business including coding. I want it to be secure and streamlined.
I have a gaming PC, it runs widows and gaming rootkits and all kinds of questionable software. I don’t want my sensitive data around that computer.
I do art on my ipad, I want it to be about art. I dont need the phone stuff there. I don’t want my personal stuff there.
> flashlight too
lol I actually carry a flashlight with me always. Phones are absolutely garbage compared to a modern High CRI flashlight. I also have a FF camera and take it with me very frequently.
You asked for it. I don't want that. I don't want an "integrated experience" and "flagship quality". The first sounds juvenile and the second sounds unnecessarily expensive and probably containing shit I don't need, like fancy cameras to look good on my nonexistent socials.
What I want is a simple, slow, old, efficient, simple phone with the interface of an 80s era 8bit computer that can actually, imagine this, make and, to complicate matters even further, even take calls.
I basically want an open source dumb phone. Do these exist? If not, why not focus on this first? Why go for fancy cameras and apps when we can't even make calls? Looking at your PinePhone.
https://en.wikipedia.org/wiki/Comparison_of_Firefox_OS_devic...
The KaiOS WhatsApp app will stop working next year though.
So the UX definitely shows complete absence of culture of design in this kind of space (eg gameboy UI design, classic dumbphone UI, ...). Small screens need very simple widgets (most importantly grid and vertical list), hand made paddings and proper space optimizations, most likely some bitmap font tailored to the resolution.. You can't just slap some dynamic layout and expect it to work nicely on such a small screen. Everything vectorized will be both costly and of dubious rendering quality, etc..
And of course these things only have "decent", but definitely not "1 month scale" battery life. I guess LTE is the real culprit here.
Your dream is achievable and has been achieved. PinePhone for example.
I obviously don't mean everyone everyone. More like the overwhelming majority of people.
To me a simple, dumb, open source device that can easily be manufactured in all kinds of conditions all over the world sounds like a dream for actual, practical purposes. Like, for example, again, calling. To some degree I have the same issues with "smart watches". Simple, open source, dumb smart watches with just a smidge of 8bit CPU goodness to display, say, something simple like a word or even a letter on the screen would be quite useful. I know there is some movement in that arena using ESP32s but I am not particularly impressed. Alas, alas. Why do we as a civilization tend to go for the extremes and not just get our basic shit together first?
I was being slightly obtuse and I understood perfectly well that you meant all reasonable people. If I can't be obtuse and pedantic on HN though, where else?
Showing how different different people are. I've probably made... 20? ish? phone calls in the last decade. (I'm 46.)
Most every phone can already do what we want but the hardware is undocumented or locked behind NDA and the firmware is hostile. Without those locks the gigacorps cant keep you in the data-mining garden. We wont have open phones until this thinking changes. Keep dreaming until then.
For 3G, you could always do that. You only needed the right modem module with voice call support and audio I/O, like bare PCM pins, and a host micro to handle AT commands.
What's wrong with the Android one - is it not permissively licensed?
I think the biggest problem of Linux phones is the community's obsession with trying (and failing) to reimplement (multiple times, in parallel) things that Android does really well and can be used as-is.
That's why the PinePhone or Librem 5 still can't even match the usability (at basic things like phone or camera or battery life) of a 2010-era Android phone, despite having similar hardware.
You want a Linux phone that actually works? Start with an AOSP-based phone and provide manufacturer-approved root and escape hatch such as first-party terminal and Wayland/X server app to run Linux apps.
Over time, you can slowly replace Android components with their Linux desktop counterparts when they're ready (or the other way around - the Android bits can just be the commonly-accepted solution to specific problems in Linux - even desktop - distros), but at least you're starting from a solid base.
It is not, AOSP based distributions have to kang it from vendor builds. Qualcomm's is mostly standardized but Samsung wrote their own stack and voLTE/voNR won't work on any custom roms.
> That's why the PinePhone or Librem 5 still can't even match the usability (at basic things like phone or camera or battery life) of a 2010-era Android phone, despite having similar hardware.
They most certainly do NOT have similar hardware. You're wrong on thinking it's a software problem when the hardware being interfaced with is notoriously proprietary. The PinePhone and Librem phones are using self-contained quectel modems connected via different interfaces. They are nothing like the integrated soc's of nearly every other device on the market. This dramatically impacts battery life and stability and I don't think it will ever be a solved problem when building devices this way.
There were pointless debates around ITU-T and 3GPP as to whether the LTE is 3.5G or 3.9G or 4G next to the pink fact elephant that _it_ is going to be _the_ next cell standard anyway. That hot debate delayed voice call discussions to post-launch matter that eventually coagulated as the VoLTE.
The Voice-over-LTE was a total kludge together that were(ARE) carrier specific implementations. In Japan at least, it seemed to have had push from KDDI, esp. with hindsight that NTT docomo and Fujitsu, both formerly influential in 3G, both seemed on fire majority of that timeframe, while Nokia being a supernova beyond fire. It would be very natural if Samsung would have made vital contributions, but I don't have much informational pressure from that direction. VoLTE is phone company mannerisms, remnants of retro-futuristic media features of 3G, weird spaghetti codes from carrier labs, and international call exchange system, all homogenized in a blender; the artefacts were co-developed pair of server and client implementations that are standardized in the way there aren't many implementations but not something carefully spec-worked before construction for interoperability. That all happened ~a decade ago. Some point between Android 4 to 10.
I suppose it was all retconned into 3GPP standards after action, but the whole stack is still like a lobotomized Android call app and private fork of Asterisk embedded into the cellular core monstrosity. I guess embedded modem people(like Quectel) had finally got to port functions into modem chip firmwares so they can make calls, after someone done it for non-smartphones(KaiOS, Smarterphone...), I think around 2019 +/- 1 year.
What I'm trying to say is, VoLTE is complicated. It's something like SIM-authenticated SIP/RTP over IP under IP, not even regular 3G data session let alone SIP-VoIP on Layer 3 UDP/IP.
Osmocom project Wiki summarizes it better than I[1]: "Voice over LTE is an adaptation by 3GPP to use IMS over an LTE cellular network. The LTE EPS (Evolved Packet System) provides the functionality of the underlying IP-CAN. ... IMS is much more than normal SIP/RTP. And in addition to that, there is a tight integration between the LTE system and the IMS on top of it."
And it's supposed to be the core feature of a phone, "not".toupper() a party trick... There are better things to do in life for most people.
Purism achieved that with their Librem 5 phone [0]. However having a polished experience like with iPhone requires to invest billions in software. It can be used as a daily driver [1] but there's a lot to improve yet. Sent from my Librem 5.
[0] https://puri.sm/posts/converging-on-convergence-pureos-is-co...
[1] https://puri.sm/posts/my-first-year-of-librem-5-convergence/
Just like DuckDuckGo I think the name will hamper adoption of this.
> Furious Support from the FuriOS team
I do like "FuriOS" and think it avoids this issue, but other portmanteaus might be better avoided.
And if you can’t understand why almost everyone will read ‘furi’ as furry and not ‘fury’, then you’ve got a phonics problem.
Curiosity got me to look.
Maybe it will drive traffic.
As such, most of these attempts aren't that attractive to me as OS nerd, always recycling the same ideas.